Data Engineering DC: Scaling Regulatory Data & Minimizing AI Misinformation

​Join us for a Data Engineering DC Meetup! We've got two killer talks this month. Come hang out, geek out, and meet some of the MOST INTERESTING people in DC's data scene!! ✨

​Talks:

• ​Talk #1: Spicy Regs - Building a Serverless Data Warehouse for Regulatory Data

  • ​Eugene Kim - Founder of KVEC.ai

• ​Talk #2: Context Engineering - Using Graphs and MCP to minimize misinformation

  • ​Jabari Grubb - Data Architect @ InPress / Founder HarborLight Nexus

​Talk #1 Abstract: How do you warehouse 30M+ regulatory comments without running a database? This talk walks through spicy-regs, an ETL pipeline that turns federal regulations.gov filings into a queryable dataset using nothing but Parquet files on Cloudflare R2 and DuckDB as the compute engine. We'll cover the file layout, Hive partitioning by agency, streaming merges that never blow RAM, and the war stories — OOMs, silent overwrites, duplicate rows — that shaped the design.

​Talk #2 Abstract: This talk introduces an architectural framework for Context Engineering for moves beyond tranditional RAG by integrating the Model Context Protocol (MCP) and Neo4j Knowledge Graphs.

​Talk #1 Speaker Bio: Eugene Kim is the founder of KVEC.ai and works on civic tech and AI-powered software, including Spicy Regs and projects with Civic Tech DC. He focuses on building practical systems that make public-interest work easier to use, share, and scale.

Talk #2 Speaker Bio: Jabari Grubb is an Independent AI Data Consultant and a Fractional Data Architect at InPress.